Would this work?

Okay, so here is a schematic that I THINK could work. My goal is a dimmable Christmas light controller. My hope is to use the daisy-chaining power of the TLC5940 to create a large number of outputs. The H11AA11 would give me a zero-cross reference, and the MOC3020 and BT139 would work together to switch the AC current.

Is this totally flawed logic? Can someone confirm that this may or may not work? Obviously there is a bunch of code involved, I’m not worried about that right now…

Also, I circled a resistor and wrote a note, “why is this here”. If someone could answer that, it would make my day. I simply saw that in another schematic that someone posted, and I can’t for the life of me figure out what it is for.

please tell me that I actually have this right…

edit: I am terribly sorry about the crappy pic attachment. I couldn’t shrink it and have it still legible, and I don’t know how to do the in-post preview that opens in a new tab when clicked…

I resampled to 1000 pixels wide. I find it legible.

Is this a schematic you found somewhere? Which pins on the Triac are MT1 and MT2?

This is a schematic that I drew, but is based on a combination of several others that I have seen in various other places.

The two center pins on the triac are the main pins, and the little one off to the top left is the control pin. Is that not the correct way to draw it?

Page 2 of the datasheet:

http://www.nxp.com/documents/data_sheet/BT139-600E.pdf

T1 should be the terminal on the bottom, and the gate should be drawn coming off of the bottom. Otherwise, the Triac is On by default.

T1 or MT1 can be thought of as the Common terminal of the Triac. If the gate is connected to T1 with the 330 ohm resistor, the Triac will remain Off.

Unfortunately, there are some websites that draw a Triac with the Gate lead coming off the top, near MT2. I find this misleading, and different from the way I originally learned to draw it. In any case, it is best to label T1/MT1 and T2/MT2 so there is no question.

I follow you now. That was what I intended to draw. I have fixed that now in my schematic. T1 and the Gate exit the bottom of the triac, and T2 is the top.

Did the rest of the circuit look solid to you? Can I couple the TLC5940 with the MOC3020 in this way? I think it should work, but in the other schematics I have seen I have only ever seen people take the MOC3020 HIGH whereas I am sinking it LOW. This is the same thing in the end, right?

Is there anything else I have majorly screwed up?

Thanks for your help!

You're putting 170V peak across the H11AA11 LED which can only withstand 5V reverse voltage.

MarkT is correct! I missed that. The H11AA11 requires a diode connected antiparallel to its LED. Antiparallel means parallel, but pointing the opposite direction.

However, then you get a single pulse only when the AC is positive polarity. A more useful way is to use a bridge rectifier at the input (LED side) of the H11AA11. In that way, the phototransistor is nearly always ON, only OFF when the AC waveform passes through zero volts, whether it is positive or negative going at the time.

As far as the input to the MOC3020, it doesn't matter if you pull it low or high, as long as it is wired so current goes the proper direction.

OK, back to the 330 ohm resistor circled- it is there to make sure that random noise picked up by the wiring, stray leakage currents, that sort of thing don't inadvertently trigger the Triac when the MOC3020 is off.

polymorph: MarkT is correct! I missed that. The H11AA11 requires a diode connected antiparallel to its LED. Antiparallel means parallel, but pointing the opposite direction.

However, then you get a single pulse only when the AC is positive polarity. A more useful way is to use a bridge rectifier at the input (LED side) of the H11AA11. In that way, the phototransistor is nearly always ON, only OFF when the AC waveform passes through zero volts, whether it is positive or negative going at the time.

optocouplers are available with anti-parallel LEDs internally, I believe...

MarkT: optocouplers are available with anti-parallel LEDs internally, I believe...

It's called a Sharp PC814 or Everlight EL814, readily available.

So life drug me away from my hobbies for a couple of weeks, but I am back at it.

What I MEANT to write above the component was H11AA1M.

The datasheet is here ---> https://www.fairchildsemi.com/datasheets/H1/H11AA4M.pdf

It SEEEMS like this model has an internal LED in BOTH directions, and therefore should work. Is this correct? Have I missed something yet again?

Thanks all!

ianahner: So life drug me away from my hobbies for a couple of weeks, but I am back at it.

I hope you were not using drugs. Other things might have dragged you away ...

ianahner: It SEEEMS like this model has an internal LED in BOTH directions, and therefore should work. Is this correct? Have I missed something yet again?

It would indeed be fine, perfect for the job just as you drew the diagram. I suppose I (nor anyone else) simply did not bother to research the component notation but went on your diagram showing only one LED in the part.

The similar part I cited may be more available and much cheaper, or do you already have the device?

For God's sake, do your testing with an Isolation Transformer so you will be alive to enjoy it at Christmas. Also, that 33K resistor in the zero cross is going to do about .43 watts, so use a 1W resistor.

Paul__B: I hope you were not using drugs. Other things might have dragged you away ...

Leave it to this forum to bring out the grammar nazis. I'll make note of my mistake and try to avoid future iterations.

Paul__B: It would indeed be fine, perfect for the job just as you drew the diagram. I suppose I (nor anyone else) simply did not bother to research the component notation but went on your diagram showing only one LED in the part.

The similar part I cited may be more available and much cheaper, or do you already have the device?

That other part may be a better option. I will look into it. The H11AA1M was simply the first solution I found that looked very promising, so I didn't dig a ton further to find alternatives. In the end, I only need one of them, so whatever I find one of cheapest that will work well will probably be the ticket for me.

rmetzner49: For God's sake, do your testing with an Isolation Transformer so you will be alive to enjoy it at Christmas. Also, that 33K resistor in the zero cross is going to do about .43 watts, so use a 1W resistor.

On the 33k resistor: that's the plan.

On the Isolation Transformer: Always a good plan if you are going to go probing around with a meter or oscilloscope. My approach was going to be just seal the whole damn thing up in a box and never open it when power was connected, but being that I do have an isotrans laying around somewhere (if I can remember where it is) it would certainly not be a bad idea to use it during the testing. Thanks for the heads up!

I'll order some pieces here in the next few days and see if I can get a working model put together. Thanks for all the help, everyone!

I suppose I (nor anyone else) simply did not bother to research the component notation but went on your diagram showing only one LED in the part.

Hey, I did. He changed to a different number.

ianahner:
Leave it to this forum to bring out the grammar Nazis. I’ll make note of my mistake and try to avoid future iterations.

Hey, but it’s fun/ funny! :smiley:

ianahner:
That other part may be a better option. I will look into it.

I mainly mention it because I have a few dozen from eBay.

That’s not to say I have used any yet …